面向智能配电网数据通信的无线传感器网络拥塞控制研究
发布时间：2018-12-17 08:19
【摘要】：具有低成本、无布线、自然隔离、易于维护特点的无线传感器网络是智能电网数据通信实践的未来发展趋势。将无线传感器网络用于智能配电网数据通信中具有很好的研究意义与应用价值。由于配电网感知事件的数据存在着突发性特点,且并发数据流互相干扰,均可能造成无线传感器网络数据通信路由路径发生拥塞,导致数据传输延时增大、丢包数量增多等诸多问题的出现,影响智能配电网安全运行。本文面向智能配电网的无线传感器网络的拥塞控制问题,对无线传感器网络实时可靠性服务质量保证机制进行了研究。本文主要研究工作如下:1、对配电自动化系统结构、智能配电网数据通信要求、无线传感器网络架构进行了分析。对智能配电网无线传感器网络的结构进行了设计,分析了造成无线传感器网络数据拥塞的主要因素。2、针对MAC层拥塞问题,将配电网数据按不同传输优先级进行分类,以建立无线传输信道不公平竞争机制。基于非线性流体流量模型,在传统的缓冲队列管理算法基础上,应用一种整体动态拥塞控制策略,设计了一种基于迭代思想的滑模学习控制器进行拥塞控制,从对参考信号的跟踪性能、丢包率和控制信号等指标与传统滑模控制器进行性能比较,并对控制效果进行讨论。3、在建立的拥塞控制策略基础上,提出MAC层性能评价体系,依据IEEE802.15.4标准MAC协议,建立相关评价网络性能的数学模型,在仿真实验中通过网络吞吐率、网络有效吞吐率、信道冲撞率和网络通信延时等性能指标,对MAC层拥塞控制方法的有效性进行评价。4、针对网络层拥塞,根据智能配电网通信性能要求的约束条件,建立了一种智能配电网无线传感器网络数据传输最优路径通信模型。利用庞特里雅金极值的方法构建出所设计智能配电网无线传感器网络通信模型的哈密尔顿函数,通过对极值条件存在的分析,判断路径是否在最优路径,最后给出了最优控制模型的求解过程。通过算例分析和性能对比仿真实验,验证所提出的方法在数据传输延时、网络传输能力以及接受数据的错误率的数据通信性能上均匀优势,对从网络层的层面缓解网络出现的拥塞是有效的。
[Abstract]:Wireless sensor networks with low cost, no wiring, natural isolation and easy maintenance are the development trend of smart grid data communication practice in the future. It is of great significance and application value to apply wireless sensor network to smart distribution network data communication. Because of the sudden characteristics of the data of perceived events in the distribution network and the mutual interference of the concurrent data streams, the routing path of data communication in wireless sensor networks may be congested and the data transmission delay will be increased. The emergence of many problems, such as increasing number of packet loss, affects the safe operation of smart distribution network. Aiming at congestion control of wireless sensor networks in smart distribution networks, this paper studies the mechanism of real-time reliability quality of service (QoS) assurance in wireless sensor networks. The main work of this paper is as follows: 1. The structure of distribution automation system, the requirements of smart distribution network data communication and the wireless sensor network architecture are analyzed. The structure of smart distribution network wireless sensor network is designed, and the main factors causing data congestion in wireless sensor network are analyzed. 2. Aiming at the congestion problem of MAC layer, the distribution network data is classified according to different transmission priorities. To establish a wireless transmission channel unfair competition mechanism. Based on the nonlinear fluid flow model and the traditional buffer queue management algorithm, a sliding mode learning controller based on iterative thought is designed to control congestion by using a global dynamic congestion control strategy. The performance of reference signal tracking, packet loss rate and control signal is compared with that of the traditional sliding mode controller. 3. Based on the established congestion control strategy, the performance evaluation system of MAC layer is proposed. According to the IEEE802.15.4 standard MAC protocol, a mathematical model is established to evaluate the network performance. In the simulation experiment, the network throughput, the network effective throughput, the channel collision rate and the network communication delay are measured. This paper evaluates the effectiveness of MAC congestion control method. 4. According to the constraints of smart distribution network communication performance, an optimal path communication model for wireless sensor network data transmission in smart distribution network is established. The Hamiltonian function of the wireless sensor network communication model of smart distribution network is constructed by using the method of Pontreyagin extremum. By analyzing the existence of extreme condition, we can determine whether the path is in the optimal path or not. Finally, the process of solving the optimal control model is given. The simulation results show that the proposed method has uniform advantages in data transmission delay, network transmission capability and error rate of receiving data. It is effective to alleviate the network congestion from the network layer level.
【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM76;TN929.5;TP212.9
本文编号：2383947
[Abstract]:Wireless sensor networks with low cost, no wiring, natural isolation and easy maintenance are the development trend of smart grid data communication practice in the future. It is of great significance and application value to apply wireless sensor network to smart distribution network data communication. Because of the sudden characteristics of the data of perceived events in the distribution network and the mutual interference of the concurrent data streams, the routing path of data communication in wireless sensor networks may be congested and the data transmission delay will be increased. The emergence of many problems, such as increasing number of packet loss, affects the safe operation of smart distribution network. Aiming at congestion control of wireless sensor networks in smart distribution networks, this paper studies the mechanism of real-time reliability quality of service (QoS) assurance in wireless sensor networks. The main work of this paper is as follows: 1. The structure of distribution automation system, the requirements of smart distribution network data communication and the wireless sensor network architecture are analyzed. The structure of smart distribution network wireless sensor network is designed, and the main factors causing data congestion in wireless sensor network are analyzed. 2. Aiming at the congestion problem of MAC layer, the distribution network data is classified according to different transmission priorities. To establish a wireless transmission channel unfair competition mechanism. Based on the nonlinear fluid flow model and the traditional buffer queue management algorithm, a sliding mode learning controller based on iterative thought is designed to control congestion by using a global dynamic congestion control strategy. The performance of reference signal tracking, packet loss rate and control signal is compared with that of the traditional sliding mode controller. 3. Based on the established congestion control strategy, the performance evaluation system of MAC layer is proposed. According to the IEEE802.15.4 standard MAC protocol, a mathematical model is established to evaluate the network performance. In the simulation experiment, the network throughput, the network effective throughput, the channel collision rate and the network communication delay are measured. This paper evaluates the effectiveness of MAC congestion control method. 4. According to the constraints of smart distribution network communication performance, an optimal path communication model for wireless sensor network data transmission in smart distribution network is established. The Hamiltonian function of the wireless sensor network communication model of smart distribution network is constructed by using the method of Pontreyagin extremum. By analyzing the existence of extreme condition, we can determine whether the path is in the optimal path or not. Finally, the process of solving the optimal control model is given. The simulation results show that the proposed method has uniform advantages in data transmission delay, network transmission capability and error rate of receiving data. It is effective to alleviate the network congestion from the network layer level.
【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM76;TN929.5;TP212.9
【共引文献】
相关期刊论文 前8条
1 刘环宇;刘强;;基于人力资源管理的企业绩效管理问题研究[J];中国商论;2017年11期
2 武漫漫;;基于HPWS理论分析中部制造企业的绩效人才队伍建设[J];中国新通信;2017年06期
3 马晓娟;;关于人力资源管理对企业绩效影响的探析[J];人才资源开发;2017年04期
4 卢海瑞;;人力资源管理与企业发展之间的相关性分析[J];大众投资指南;2017年01期
5 闫美灵;;基于因子分析的人力资源管理与企业绩效关系研究[J];科技创新与生产力;2016年12期
6 许殷莉;;人力资源管理实践差异与员工关系影响分析[J];商场现代化;2016年25期
7 濮锐;江衷;;事业单位绩效改革成效及其对员工工作倦怠的影响研究[J];中共浙江省委党校学报;2016年05期
8 谌珂;;基于人力资源管理的企业绩效研究[J];商业故事;2016年06期
，本文编号：2383947
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2383947.html
